upstream-archive/presidio
1
0
Fork 0
mirror of https://github.com/data-privacy-stack/presidio.git synced 2026-09-21 13:38:05 -05:00
Code Issues Packages Projects Releases Wiki Activity

Changed default aggregation_strategy to max (#1342)

Browse Source
This commit is contained in:
Omri Mendels
2024-03-25 11:19:25 +02:00
committed by GitHub
parent db8ff82541
commit 49a996de8b
2 changed files with 7 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
presidio-analyzer/conf/transformers.yaml
+6 -6
View File
@@ -9,12 +9,12 @@ models:
ner_model_configuration:
labels_to_ignore:
- O
aggregation_strategy: simple # "simple", "first", "average", "max"
stride: 16 # If stride >= 0, process long texts in
# overlapping windows of the model max
# length. The value is the length of the
# window overlap in transformer tokenizer
# tokens, NOT the length of the stride.
aggregation_strategy: max # "simple", "first", "average", "max"
stride: 16 # If stride >= 0, process long texts in
# overlapping windows of the model max
# length. The value is the length of the
# window overlap in transformer tokenizer
# tokens, NOT the length of the stride.
alignment_mode: expand # "strict", "contract", "expand"
model_to_presidio_entity_mapping:
PER: PERSON
presidio-analyzer/presidio_analyzer/nlp_engine/ner_model_configuration.py
+1 -1
View File
@@ -64,7 +64,7 @@ class NerModelConfiguration:
""" # noqa E501
labels_to_ignore: Optional[Collection[str]] = None
aggregation_strategy: Optional[str] = "simple"
aggregation_strategy: Optional[str] = "max"
stride: Optional[int] = 14
alignment_mode: Optional[str] = "expand"
default_score: Optional[float] = 0.85